CEN Joint Projects

Back to Joint Projects List

Effect of shrubs on snow mantle thickness and thermal regime in Arctic tundra

Kinnard, arbuste_neige, BigFormat

Photo Credit: Vincent Houde

Research Team



Global warming is leading to an increase in the cover and size of shrubs in the Arctic, a phenomenon known as Arctic shrubbery. Shrubs modify the interactions between the atmosphere and permafrost in a number of ways, particularly in winter by intercepting wind-blown snow. The preferential accumulation of snow around shrubs promotes the thermal insulation of permafrost and thus its warming, which can accelerate microbial decomposition and the release of greenhouse gases from the soil to the atmosphere. Recent studies have also shown that the branches of shrubs can remove heat from the ground in winter, counteracting the insulating effect of snow cover. Very few spatially resolved measurements have been collected to investigate the interactions between shrubs and snow cover in the tundra. This project aims to fill this knowledge gap by using UAVs to map snowpack thickness and temperature, as well as shrub distribution and structure, in tundra shrublands on Bylot Island, Nunavut. These measurements will be coupled with snowpack temperature monitoring inside and outside the shrublands to investigate the impact of shrublands on snow and underlying permafrost temperatures.

The results will show whether and how the spatial distribution of snow cover is affected by the size, structure and distribution of shrubs in the landscape, and whether there is a warming or cooling effect.

Progress (Year 1- April, 2024)

The joint project on the effect of shrubs on the thickness, temperature and spatial distribution of snow cover in the Canadian High Arctic is currently in the data analysis phase.

The field campaign took place from May 11 to June 29, 2023 in the Quarlikturvik valley on Bylot Island, Nunavut. The campaign went smoothly despite a late snowmelt that caused some minor complications, but the pre-departure objectives were largely met. Several drone flights were conducted before, during and after the melt. Two measuring stations (shrubs vs. prostrate vegetation) were set up, equipped with temperature and humidity sensors (see photos in the appendix). Automatic cameras were also installed, with graduated poles to measure snow height during the winter period. These temperature and humidity sensors and cameras will be recovered in the summer of 2024.

Unfortunately, some technical problems beyond our control, mainly related to the drone data, have delayed the data processing. We are currently conducting tests to find the best possible classification method for discriminating between shrubs and bare ground, which is challenging given the high micro-topographic variability of the ground and the small size of the shrubs (<60 cm) (see preliminary results presented at the CEN symposium). Much of the work to date has therefore been devoted to testing different classification methods and software.

Finding a method to classify a photogrammetric point cloud in the presence of low vegetation is a methodological challenge that has received little attention in the scientific literature. Our project will at least address the question of the performance and viability of photogrammetry in such a situation, and compare the solutions we obtain with what is done with LiDAR.

Once we have obtained a classification that is considered optimal, digital terrain and canopy height models will be generated, which will then be used to generate snow height maps. These maps will be analyzed to determine the relationship between snowpack height, ground topography, and shrub presence and size at a relatively large scale (3.8 square kilometers).

Despite the technical problems encountered during data processing, the project has great potential in terms of the results that can be obtained in the coming months. During the 2023 field campaign, a large amount of data was collected (photogrammetry, temperature station, snow height camera, terrestrial LiDAR survey, shrub mapping), and this variety of data will lead to many scientific spin-offs, even beyond the current project.

List of Communications

  1. Scientific poster - Colloque CEN, Université Laval, 2024 (See Appendix below)
  2. Publication in preparation: Comparing spectral albedo and NDVI of herbaceous and shrub tundra vegetation at Bylot Island, Canadian High-Arctic: Florent Domine, Maria Belke-Brea, Ghislain Picard, Laurent Arnaud, Esther Lévesque, Arthur Bayle, and Christophe Kinnard.
UAV data collected and manual shrub surveys collected in 2023 were used to test whether erect shrubs have a spectral signature that allows them to be mapped. We find that shrubs cannot be discriminated spectrally, which reinforces the need to develop mapping methods from 3D point clouds, something Vincent Houde is currently doing as part of his master's degree and this joint project. . This article is about to be submitted to Biogeoscience.

Scholarship Supplement

Congratulations to Master's student Vincent Houde, who has been awarded the $5,000 CEN scholarship for the 2023/2024 academic year.

Photo Gallery

© 2024 Centre for northern studies - All rights reserved